tomcat-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Johnny Kewl" <j...@kewlstuff.co.za>
Subject Re: Server Node Id
Date Thu, 28 Jun 2007 10:03:07 GMT
Also, maybe this is a good idea, one other thing I found handy behind a JK 
cluster is leaving 2 connectors open.
So you got the cluster server doing its round robin, but each machine can 
find the other machine directly through say port 8080.
So again in the dB, you just have to put the address of all the machines in.
Then if one machine say gets a file upload, it sends that to all the other 
machines, but through the direct port to bypass the clustering.

Maybe.... sorry couldnt help you more


----- Original Message ----- 
From: "Ingo Düppe" <mailing-lists@dueppe.com>
To: "Tomcat Users List" <users@tomcat.apache.org>
Sent: Wednesday, June 27, 2007 10:13 PM
Subject: Re: Server Node Id


> Well, the reason is, that I need to run some tasks that have to be 
> executed on each node once within the cluster. So each node must be able 
> to identify these tasks that still need to be executed on this node. So in 
> my solution the server node gets its id and checks which was the last task 
> it has executed.
>
> Therefore the id must not change after each startup.
> - Ingo
>
> Johnny Kewl schrieb:
>> Interesting question... I dont know if there is a property one can call.
>> I had a similar need behind a JK load balancing scheme, what I did was 
>> just generate a unique ID when each cluster webapp starts up, something 
>> like the MS GUID idea.
>> No pre-configuration required.
>>
>> I think it would be very interesting to know why you doing this, think 
>> the guys will then really be abe to give you some good idea's.
>>
>>
>> ----- Original Message ----- From: "Ingo Düppe" 
>> <mailing-lists@dueppe.com>
>> To: "Tomcat Users List" <users@tomcat.apache.org>
>> Sent: Wednesday, June 27, 2007 12:07 PM
>> Subject: Server Node Id
>>
>>
>>> Hello,
>>>
>>> i run an application within a tomcat cluster. Now I need a unique key 
>>> for each server node for the application. My first solution was a 
>>> property file within the war file. But this is not so handy on 
>>> deployment cycles.
>>>
>>> Is there a better way to place a unique identify for each cluster node 
>>> or does a identifier already exist.
>>> The id must only be unique within the cluster not globaly.
>>>
>>> Regard
>>> Ingo
>>>
>>> ---------------------------------------------------------------------
>>> To start a new topic, e-mail: users@tomcat.apache.org
>>> To unsubscribe, e-mail: users-unsubscribe@tomcat.apache.org
>>> For additional commands, e-mail: users-help@tomcat.apache.org
>>>
>>>
>>
>>
>> ---------------------------------------------------------------------
>> To start a new topic, e-mail: users@tomcat.apache.org
>> To unsubscribe, e-mail: users-unsubscribe@tomcat.apache.org
>> For additional commands, e-mail: users-help@tomcat.apache.org
>>
>>
>>
>
>
> ---------------------------------------------------------------------
> To start a new topic, e-mail: users@tomcat.apache.org
> To unsubscribe, e-mail: users-unsubscribe@tomcat.apache.org
> For additional commands, e-mail: users-help@tomcat.apache.org
>
> 


---------------------------------------------------------------------
To start a new topic, e-mail: users@tomcat.apache.org
To unsubscribe, e-mail: users-unsubscribe@tomcat.apache.org
For additional commands, e-mail: users-help@tomcat.apache.org


Mime
View raw message